Geography and Vision

Seeing, Imagining and Representing

the World

Denis Cosgrove
 9 Moving maps

The historiography of maps and mapping – of ‘cartography’, to use the

nineteenth-century neologism – has been revolutionized in the past
quarter-century. The shift is as profound as the changes that have taken
place in map-making and map use themselves, resulting from satellite
remote sensing, digitization of spatially referenced data and computer
manipulation of information and images. Three main changes in carto-
graphic historiography are worth highlighting. First, there has been detailed
exposure of the normalizing and often ideological authority of maps, and
criticism of the active roles cartography has played in the nexus of power-
knowledge that frames and shapes the geographies of the modern world.
Geographic and topographic mapping and maps have been critical tools for
the modern state and its agencies in shaping social and moral spaces, and
they played a central role in the Western physical and intellectual coloniza-
tion of territories, peoples and the natural world. For over two centuries
thematic and statistical maps have extended these roles in supporting the
bureaucratic concerns of the modern state. Historians of cartography have
examined these processes across many specific instances.1 Second, map-
making’s scientific claims to offer progressively accurate and objective,
scaled representations of spatial relations, have been challenged with recog-
nition of the inescapable imaginative and artistic character of cartographic
process and products that accompany framing, selection, composition and
graphic representation of mapped information. Colour and symbolization,
for example, are chosen and applied to maps according to widely accepted
design principles, but their relationships to the appearance of landscapes
represented on topographic maps, or the bands within the infrared spec-
trum on a remote sensed image, are necessarily arbitrary. This recognition
has opened up an exciting new field of connections between scientific map-
making and creative art practices.2 Third, and closely related to the first two
developments, is the recognition of mapping as a complex cultural process

155
 GEOGRAPHY AND VISION

in which the map itself represents merely one stage. To understand the
contents, meaning and significance of any map requires that it be reinserted
into the social, historical and technical contexts and processes from which it
emerges and upon which it acts. This involves examining the map not only
as a discrete object but as the outcome of specific technical and social
processes and the generator of further social processes as it enters and
circulates in the social world.3
These assumptions of the critical approach to the nature and history of
maps are today widely accepted among historians and writers on carto-
graphy, but many would argue that they are still insufficiently embraced by
practising map-makers, especially today when many of those producing and
manipulating maps are not trained cartographers. Most professional carto-
graphers are acutely aware of the limits of their art even as they strive for
disinterested objectivity and scientific integrity in their map-making.4 But
maps made by formally trained cartographers constitute an ever-smaller
proportion of the map images available today, especially those available
online rather than drawn or printed. Geographic Information Science (GIS),
working with remote sensed digital data at intensive scales and across a
colour spectrum that stretches deep into the infrared regions, generates a
vast range of virtual and actual cartographic images. Their makers’ primary
training and interests are often in information technology and its applica-
tions to geo-referenced data rather than in the conventional cartographic
techniques and operations of projection, compilation and selection, fram-
ing and design.5 The sheer technical wizardry and compelling graphic
effects of such animated mapping packages as Google Earth offer an illusion
of total synopsis and truthful vision, and can easily blunt critical responses
and obscure significant continuities in cartographic culture.6 In this chapter
I explore some of those continuities, drawing on the critical cartographic
literature to comment on aspects of maps and mapping practices that can
easily become obscured in our excitement with the technical advances that
have made ‘mapping’ such a dynamic contemporary field of practice and
study.

Directions in critical cartography

Critical study of cartography can proceed from two directions: either
through study of the finished map – judging its function, technique, aes-
thetics and semiotics; or through a study of mapping processes, convention-
ally grouped under the headings of survey, compilation and design. From
the first perspective we might consider Abraham Ortelius’s Theatrum Orbis
Terrarum of 1570 (Fig. 9.1). Functionally, this well-known historical map
provided what is considered the first modern atlas with an opening image

156
Figure 9.1 Abraham Ortelius, Typus Orbis Terrarum, 1570 (UCLA Library, facsimile copy)
 GEOGRAPHY AND VISION

of the terraqueous globe according to the most recent information available

at the time of its making.7 The search for empirical truth is apparent from
the second edition of the map, made a mere decade later when, among
other changes, the shape of South America is more accurately portrayed.
Ortelius’s selection of individual colours for the continents anticipates their
representation on succeeding continental maps, and the ordered summary
of geographical knowledge that constitutes the atlas. Technically, the world
map uses Ptolemy’s second projection, extending the meridians to show the
whole southern hemisphere. The map is thus centred on the Equator, with a
prime meridian running through the Azores, curving the longitudes
towards the poles. Like any projection of the sphere, this has distorting
effects on shape and direction. The oval planisphere is framed with clouds
that represent the element of air, but otherwise it is relatively free of decor-
ation, apart from the title cartouche and a lower banner containing a Latin
sentence attributed to Cicero. The map offers a memorable and uncluttered
image of the globe’s lands and seas to which subsequent maps in the atlas
can be related. Yet in ways that are not immediately apparent on the map’s
surface, aesthetics could be said to trump scientific knowledge in the
balancing landmasses north and south of the known continents: remnants
of philosophical and religious belief in a harmonious distribution of lands
over the earth’s surface. The semiotics of the map are as significant as its
scientific, technical and aesthetic aspects. The text at the base of the map, for
example (which in the second edition is reinforced by four other passages
from Cicero and Seneca), reads ‘For what can seem of moment in human
affairs for him who keeps all eternity before his eyes and knows the scale of
the universal world?’ It reminds us that in the sixteenth century the world
map played a role beyond that of scientific instrument and artistic image; it
was a moral text reminding the viewer of the insignificance of human life
compared to the vastness of creation.8 In presenting the mapped ‘theatre of
the world’ (the title of Ortelius’s atlas) as a moral space, the map itself gains
an emblematic quality.9 This aspect of mapping can be traced in the West
back to the medieval Christian mappae mundi, and is a common feature of
non-Western cartography too.10 Indeed it has never disappeared from carto-
graphic culture.
The alternative approach to a critical understanding of the cartographic
image is via an examination of survey and compilation, and these will be
my focus here. By survey I mean the direct collection and production of the
spatial data to be represented, or ‘mapped’. This includes both the spatial
calculation used to create a base map, for example a local traverse or a
geodetic measure, and the informational content to be represented.11
Survey, or reconnaissance, has traditionally been a field-based activity,
within which instrumentation has played a formative role. By compilation I
mean the gathering together of surveyed information at a single location –

158
 MOVING MAPS

the cartographer’s office, laboratory or studio – and its technical transform-

ation into the finished map object. This approach privileges the process of
mapping over the product, the map, and it has attracted increasing attention
in recent years, especially as its processes impact upon the knowledge
claims that might be made of the resulting map.12 Of particular concern
have been the various means whereby knowledge gained in survey is trans-
ferred back to the place of compilation, and whereby the map itself enters
recursively into circuits of knowledge that generate further mappings.13
Mapping is itself a spatial process that involves negotiating various aspects
of securing and maintaining the integrity of cartographic data as they
circulate in space.

Survey
Survey is an embodied process involving direct, sensual contact with the
spaces to be mapped. The distanced sense of vision is privileged here, as the
word itself implies (Latin: super-video), although other senses can be critical
in specific mapping situations, such as the whole body engagement
involved in plotting precise locations or fixing trigonometric survey points
in the field, or in negotiating darkened spaces for underground mapping,
where bodily touch is more significant than sight.14 Historically, there has
been a progressive shift away from the individual human body as a reliable
agent for recording spatial information, towards dependence upon instru-
mentation as the guarantor of accuracy and objectivity in survey data. This is
apparent in the use of compass, astrolabe and cross staff, later alidade and
lens-based instruments, and most recently light-sensitive remote sensing
aids to the human eye. Optical instruments not only extend the scope of
human vision; historically they have been used to supplant it. Thus
Galileo’s revolutionary mappings of celestial movement and imperfections
on the surface of celestial bodies were founded on the capacity of the
telescope lens not only to reveal the surface corrugations of the moon
(traced directly onto paper by Galileo’s hand), but to allow the sun to burn
the pattern of its dark spots directly onto the paper with no apparent human
intervention.15 The eighteenth century saw the radical extension of instru-
mentation to all aspects of reconnaissance: the alidade and plane table, as
well as accurate geodetics, made possible the production of national maps
based on triangulation, a technique that had been theorized over 200 years
earlier by Gemma Frisius. In the same years the mercury barometer was
used to measure and plot altitude, while accurate mapping of oceanic space
only became possible with John Harrison’s 1780s invention of the chron-
ometer, which allowed relatively easy and secure measurement of longitude
at sea.16

159
 GEOGRAPHY AND VISION

But despite increased reliance on instrumentally measured survey, the

human eye has remained a crucial element of mapping and the use of
maps. Into the twentieth century, the British Admiralty required its
officers, including the most junior, to learn accurate sketching as a means
for gathering and recording information about coastlines and harbours,
considering drawing to be superior to any written account. Sketches were of
two types: the memorial sketch, ‘a delineation of a harbour, or any part of a
coast, from the memory only, without notes or any immediate sight’, con-
veying ‘the general area of a bay, harbour, or island . . . shewing that some
such places are there’, and the eye-sketch, ‘done by the eye at one station,
without measuring distances; and drawn according to the apparent shape
and dimensions of the land’.17 The invention of photography and the use of
balloons, followed by powered flight, furthered the displacement of the
human eye in geographic and topographic mapping. In 1915 the German
Oskar Messter invented an airborne automatic camera that could film a
60km by 2.4km strip of the earth’s surface in a sequence of overlapping
frames, to be either printed as they were or used as raw visual data. Aerial
photography thus replaced to some degree the epic work of Cassini or
Everest in surveying great arcs of meridian from Dunkirk to Perpignan and
Bangalore to Delhi respectively.18 Photogrammetric survey was used in
mapping great colonial stretches of Africa, Australia and Antarctica into the
1950s. But even aerial photography and its contemporary successor, remote
sensed imaging from orbiting satellites far above the earth’s surface, have
not wholly replaced the sensing human body. ‘Ground truthing’ was crucial
for removing the errors caused to the 1950s British Antarctic Survey by
magnetic deviation, cloud cover and distance distortions in polar regions.19
It remains necessary to ensure the instrumental accuracy of remote sensed
maps today.
In the mapping process, the increased accuracy and consistency of survey
results secured by the replacement of the sensing but subjective human
body by instruments are always threatened by the problems of transferring
recorded data from the place of survey to the place of compilation. The
sketch-map can play a role in this, but as the name suggests, it lacks the
authority of the ‘true’, surveyed map. The issue of securing the accuracy of
mobile knowledge is beautifully expressed in Le Petit Prince by the French
writer (and early denizen of aerial survey in French colonial Africa), Anto-
ine de Saint Exupéry.20 His eponymous hero flies from planet to planet
absorbing moral lessons from their inhabitants and discovering the strange
habits of the adult world. The most beautiful of all the planets the Little
Prince visits is occupied by a single old man, seated at a desk and inscribing
information into a great book. He is a geographer, but he claims never to
have seen the beauty of his planet. Bemused, the Little Prince asks why. The
geographer responds as follows:

160
 MOVING MAPS

‘The geographer is not the one who counts the towns, rivers, moun-
tains, seas, oceans and deserts. The geographer is much too important
to go wandering about. He never leaves his study. But he receives
explorers. He interrogates them and notes their records. And if the
records of one of them seem interesting the geographer makes an
enquiry into the explorer’s moral character.’
‘Why is that?’
‘Because a lying explorer would have catastrophic consequences for
geography books [read maps]. The same is true for an explorer who is
a drunkard.’
‘How come?’ said the Little Prince.
‘Because drunkards see double. So the geographer would note down
two mountains where only one exists.’21
I shall return below to this question of securing the truth of survey
knowledge as it travels over space.
Careful instrumentation, highly regulated recording procedures, and
learned sketching techniques have all been deployed to overcome the
subjectivity of embodied observation, but its removal is never complete.
Cartographic instruments themselves have to be tested and calibrated: James
Cook’s Pacific navigation was partly intended to test Harrison’s chron-
ometer. And during the seven-year survey by Pierre Méhain and Jean-
Baptiste Delambre, begun in 1792 to calculate the precise length of an arc of
meridian in order to determine the length of the metre as universal measure
(determined objectively as one forty-millionth of the earth’s polar circum-
ference), ‘every page of the expedition’s record was signed by each member
of the expedition or by outside witnesses to certify that the recorded meas-
urements had been performed as described. No subsequent changes were
permitted . . . the signatures radically transformed the status of the
document.’22

Compilation

Compilation brings its own interruptions to the apparently smooth transfer

of spatial information from the territory to the map. Reflecting anxiety in
some measure over these problems, the history of cartographic design has
conventionally been told as a transition from art to science, a progression
from the pictorial style that we associate above all with baroque maps, to the
unornamented ‘plain style’ of graphic presentation in the eighteenth
century. Evidence for this progression was to be found in the changing
appearance of maps themselves: the removal of cartouches, elaborate letter-
ing and extraneous information and marginalia, and the systematic use of

161
 GEOGRAPHY AND VISION

non-pictorial cartographic signs. Removal of such pictorial elements works

alongside other technical devices: the graticule and grid as the map’s con-
trolling spatial metrics, the removal of hachures and relief shading in favour
of measured contours and spot heights to indicate topographic relief, the
appearance of mathematical and numerical information (compass bearings
and geodetic information, dates of survey and publication, scale, etc.) in the
map’s margins, and the insertion of a formal key, explaining and control-
ling the use of colour and symbols. All these act as much to demonstrate the
scientific credentials of the map’s compilation as to provide for its actual
use. But we know that many such compilation decisions are inevitably arbi-
trary or driven by quite other than scientific considerations: why should
water be coloured blue rather than green or turquoise, or lowlands coloured
green or roads red? Why does the United States Geodetic Survey’s topo-
graphic map indicate schools, but not the religious denomination of cemet-
eries or the population size of municipalities indicated on the topographic
maps prepared by the French Institut Géographique Nationale? Why do
British Ordnance Survey maps mark and differentiate archaeological sites by
Roman and Gothic lettering? Even the remote sensed image is a product of
colouring choices applied by the map-maker to pixels received by the
cartographic studio in numerical, digitized form, as is apparent when one
moves across the virtual surfaces of Google Earth. The colouring of the
1997 map of ocean temperatures that made the El Niño phenomenon so
graphically compelling was inspired by the desire for graphic impact rather
than scientific objectivity.
The intimate relations between mapping and science were initially forged
in the late eighteenth and early nineteenth centuries. Enlightenment passion
for universal measure and objective precision found expression in statistics,
and statistics had their greatest social impact through graphic expression in
graphs, charts and maps. The thematic map is an invention of this historical
moment. Although William Playfair had pioneered ‘lineal arithmetic’, Alex-
ander von Humboldt is credited with producing the first isoline map in
1817 (Fig. 9.2). From the mid-eighteenth century, the shaded, or choropleth
map, which uses territorial boundaries as containers for scaled statistical
observations, replaced the tradition of recording numbers directly onto
the map. Statistical maps commanded widespread respect as a vehicle for
demonstrating causal connections between spatially correlated phenomena.
As conventionally told, it was the medico-statistical mapping of cholera
patterns in mid-nineteenth-century European cities, especially Edwin
Chadwick’s and John Snow’s 1840s maps, that secured the scientific status
of statistical mapping (Fig. 9.3). But as Tom Koch’s historical analysis of this
story insists, disease maps acted as propositions rather than scientific repre-
sentations and their scientific status should be regarded as such.23 Nonethe-
less, while the topographic map represents the specificity and uniqueness of

162
 MOVING MAPS

Figure 9.2 Heinrich Berghaus, world map showing isotherms, based on

Alexander von Humboldt, 1845

places, the thematic map is a tool for nomothetic thinking and thus for
social planning. Dot maps, for example, were especially popular among
medical doctors committed to the neo-Hippocratic thesis of environmental
causes for disease, until Pasteur’s work refocused attention on internal and
biological causes.

Thematic mapping and science

The authority of thematic mapping derives from the statistical foundations
of the information it conveys. Given that such maps are generally produced
by agencies for the better management of local territories, the problems of
transporting knowledge are often not as great as they are with exploratory
geographic or topographic mapping. But thematic mapping suffers from
two fundamental and often unacknowledged weaknesses, beyond the obvi-
ous methodological and design problems of interval and scaling choices.
First, the spatial correlations which such maps suggest can too readily be
interpreted as causal: what is called the ‘ecological fallacy’. An example of

163
 GEOGRAPHY AND VISION

Figure 9.3 Cholera map of Exeter, 1832 (Bodleian Library, Oxford)

this is the nineteenth-century medical hypothesis that high olive oil con-
sumption might be the cause of hiatus hernia, a belief based on an assumed
spatial correlation between the incidence of the disease and olive oil within
the Midi diet. It was Malgaigne’s 1840 Carte de la France hernieuse (‘Hernia map
of France’) that undermined the claim, illustrating the incidence of hernias
among the population of French departments by shading at six intervals the
numbers of people per ‘hernious’ individual within each administrative
unit, and then superimposing dietary boundaries to reveal possible causes
of the variation.24
The second weakness of the choropleth map as a scientific tool is icono-
graphic: the choice of colour shading to illustrate interval differences. The
concept of ‘shading’ itself carries powerful moral connotations; this was
especially true in an era of self-conscious ‘enlightenment’, when darkness
and shadow implied ignorance and decay, both physical and moral. Thus

164
 MOVING MAPS

Edward Quinn’s 1830 Historical Atlas used a similar device to Abraham

Ortelius’s parting clouds to surround his images of the known world at
different stages in human history. But in Quinn’s case they reveal the
darkness of ignorance being pushed progressively aside by the onward
march of civilization through the ages. On nineteenth-century thematic
maps, colour shading consistently used dark tones to register failure in
the spatial narrative of progress. Thus French statistical cartography
throughout the century, in maps of educational standards or poverty and
social provision, regularly divided the country into two parts: an ‘enlight-
ened’ north where lighter shades dominated, and an ‘obscure’ France
south of a line from San Malo to Geneva, whose failures were dimly
visible through the cartographic gloom. A similar use of shading and the
ecological fallacy is to be found on what today appear extraordinary
choropleth maps published in scientific geographical journals to illustrate
correlations between the distribution of ‘climatic energy’ and that of
‘civilization’ (Fig. 9.4).
If statistical mapping lost favour among medical researchers in the later
nineteenth century, its popularity among social planners and ‘hygienists’
peaked in the early years of the twentieth century, as the ‘civilization’
maps suggest. Statistical survey and mapping of what were deemed cul-
tural traits, such as language, dialect and custom, were central to the
arguments of supporters of volk and nation in determining the territorial

Figure 9.4 Ellsworth Huntington, World isoline map showing levels of

‘civilization’ (Mainsprings of Civilization, 1945; author’s copy)

165
 GEOGRAPHY AND VISION

extent of the unifying German Reich in the 1850s and 1860s.25 From the
work of social statisticians and statistical cartographers such as Frédéric Le
Play in France, Henry Mayhew in Britain, and Heinrich Berghaus in
Germany, and American geological and soil surveyors such as John Wesley
Powell or Eugene W. Hilgard, emerged the idea of mapping as a critical
tool of state policy in the ‘Progressive Era’ at the turn of the twentieth
century. The concept of the survey was transferred from the exploration
and inventory of colonized lands to hidden features of metropolitan soci-
eties. In the minds of social philosophers such as Patrick Geddes, a unified
survey of the physical and social characteristics of a region could provide
an ecological portrait of community in place, a foundation for its rational
planning, and a stimulus to the civic virtue and community participation
of its people.
The survey did not require exploration of unknown spaces, but an
‘archaeology’ of interconnected relations within a place that could be
revealed by the very process of mapping. Such mapping would most
appropriately be undertaken by the community itself, working under
instruction, and in the process contributing to the further consolidation of
civic virtues. Public display of the resulting maps would further this goal.
Thus, in the United States, various states and counties in the 1920s, espe-
cially in New England and the Mid-West, encouraged schoolchildren to
participate in local surveys that would reveal the true state of their com-
munity, especially regarding its physical and moral heath. These maps had
the effect of moral self-regulation, as in the example of a publicly displayed
Springfield, Massachusetts survey map that flashed 1250 coloured lights
indicating the distribution of babies born in 1913, distinguishing between
green lights for homes where the birth was registered, and red for unregis-
tered births (presumably out of wedlock). In Britain, the survey movement
peaked in the 1930s with the recruitment of schoolchildren from across
England and Wales to map the use of every parcel of land, returning the
results to London where they were compiled onto topographic base maps
and coloured to reveal national patterns of land use. Deep reds and purples
dramatically revealed the octopus sprawl of great cities such as London
whose suburban anonymity was supposedly threatening traditional civic
virtues26 (see Plate 6).
The authority of survey mapping peaked in the post-war years of welfare-
state planning and modernist social engineering, which depended heavily
on map overlay techniques both to analyse social patterns and ‘problems’
and to develop persuasive arguments for their solution. The ‘plan’, in which
the map played as significant a role as the written text, became a central
instrument of policy. Even as comprehensive spatial planning came under
serious criticism in the late 1960s and early 1970s, the value of
the cartographic techniques upon which it depended was strongly reas-

166
 MOVING MAPS

serted in ecological landscape design. The influential Scottish writer Ian

McHarg recommended the use of polygonal overlays to connect various
landscape features:
by starting with bedrock geology and then surficial geology, and then
reinterpreting these to reveal groundwater hydrology, you explain
physiography and also surficial hydrology. This then leads you inevit-
ably to soils, which leads you to plants, which leads you to animals,
which can lead you to land use. . . . every one of these steps is in fact
either correcting or reinforcing.27
McHarg is commonly credited with pioneering the cartographic tech-
niques now computerized within Geographical Information Science. GIS
draws upon and synthesizes a range of spatial representations – the aerial
photograph, remote sensed image and topographic map – as well as the-
matic spatial statistics to generate complex overlying patterns, enhanced and
manipulated by the computer. The resulting images have enormous graphic
power, reshaping our vision and understanding of the world and our
capacity to intervene in its material and social processes. Combined with the
capacity to transmit data instantaneously and manipulate it in real time on
the PC screen, we have hugely increased the cartographic illusion of syn-
optic vision and action at a distance (the magic of maps). But we need to
remain as alert as ever to the persuasive power of these newer cartographic
images. The widely reproduced NASA image of ‘Spaceship Earth’ that I have
referred to already in these essays, and whose authority stems from its
photographic ‘realism’, has been read universally but uncritically as a sign
of a vulnerable globe threatened by anthropogenic environmental crisis,
and thus as a moral mapping of human treatment of a nurturing mother
earth. In fact the image itself contains no evidence that necessarily supports
such a claim. The Apollo photograph, like the nineteenth-century disease
map or a twenty-first-century GIS image, is propositional as much as it is
representational.28

Mapping and circulating knowledge

The story of how the whole-earth image was obtained by the Apollo astro-
nauts, returned to earth, circulated globally, interpreted and gained agency
in the discourses of environmentalism and one-world globalization,
exemplifies many of the continuities in mapping as a spatial process.29 In his
work on the making and circulation of scientific knowledge, Bruno Latour
has used the term ‘immutable mobile’ to characterize those material agents
that permit scientific discourse to sustain its claims of empirical warranty
and repeatable truth in the absence of eyewitness evidence.30 The map is a

167
 GEOGRAPHY AND VISION

perfect exemplar of the immutable mobile: a container of information gath-

ered at specific locations, returned to a ‘centre of calculation’, and then
placed once more into circulation as a vehicle and instrument of scientific
knowledge and further hypotheses. The entire history of cartography can be
told as a history of struggle to realize such a status for the map. Thus
Claudius Ptolemy, whose tables of locational coordinates introduced mod-
ern mapping to Renaissance Europe, may never have drawn actual maps. His
book gave sufficient information for a skilled reader to construct a projec-
tion and plot the coordinates necessary to produce the maps from his data.
Text and tabulated figures are far more easily and accurately copied and
transported than a set of drawn maps. Securing the immutability of the
mobile has been a constant obsession of cartography. It is fundamental to
the map’s claim to be more than an imaginative picture. Indeed, carto-
graphers have actually drawn upon the authority of cartographic procedure
to grant legitimacy to what were in fact complete fabrications. Thus the
sixteenth-century French cosmographer André Thevet plotted lines of
latitude and longitude around maps of completely illusory islands.31 Such
charlatanry reveals the ultimate impossibility of the cartographic conceit.
The only true map is the territory itself, as Louis Borges long ago pointed out.
The search to secure immutable mobility for the map reveals another
feature: cartography’s prosthetic quality. The map is one of those instru-
ments that serves to extend the capacities of the human body. Like the
telescope or microscope, it allows us to see at scales impossible for
the naked eye to see and without moving the physical body over space. The
thematic map reveals the presence of phenomena that are beyond our nor-
mal bodily senses, as for example a trend surface map of property values or
of air pollution. The map also has a powerful recursive quality, at once a
memory device and a foundation for projective action. This is immediately
apparent in European mapping in the ‘age of discovery’, where the map was
at once a necessary starting point for the exploration process and a principal
outcome of that process.
These prosthetic and circulatory aspects of mapping are true also for the
social survey, and they remain so for the most technically advanced map-
pings of today. It is these features of the mapping process that makes it such
a fertile and powerful epistemology in knowing and representing the world.
The map is at once empirically rooted and imaginatively liberated and liber-
ating. No spaces can be controlled, inhabited or represented completely. But
the map permits the illusion of such possibilities. Mapping is a creative
process of inserting our humanity into the world and seizing the world for
ourselves. This is why today the boundaries between the art and science of
mapping, so long and so arbitrarily surveyed, charted and policed, are
increasingly smudged and faded, and why the imaginative and projective
potential of mappings has become so vitally present in contemporary life.

168